Molecular Weight Dependence of Lamellar Structure in Styrene

(1) Jaacks, V.; Mathes, N. Makromol. Chem. 1970,131, 295. T.; Kobayashi, T.; Chow, T. Y.; Kobayashi, S. Ibid. 1979,12,. (2) Mathes, N.; Jaacks, V. Mak...
0 downloads 0 Views 2MB Size
Macromolecules 1982, 15, 258-262

258

starting line as a somewhat dispersed shape. All polymer IV molecules have, therefore, a negative charge, whereas the polymer (A) molecules are macrozwitterions. These results lead to the conclusion that transfer or termination reactions did not occur during the polymerization.

Acknowledgment is made to the Ministry of Education for generous financial support of this work. References and Notes Jaacks, V.; Mathes, N. Makromol. Chem. 1970,131, 295. Mathes, N.; Jaacks, V. Makromol. Chem. 1971, 142, 209. Jaacks, V.; Franzmann, G. Makromol. Chem. 1971,143,283. Ludwig, E. B.; Gantmacher, A. R.; Medvedev, S. S. Symp. Macromol. Chem., Wiesbaden 1959, Paper IIIA, 12. (5) Kern, W.; Jaacks, V. J. Polym. Sci. 1960, 48, 399.

(1) (2) (3) (4)

(6) Jaacks, V.; Kern, W. Makromol. Chem. 1963,62, 1. (7) Bown, C.; Ledwith, A.; Matheis, P. J . Polym. Sci. 1959,34,93. (8) Yamashita, Y.; Okada, M.; Suyama, K. Makromol. Chem. 1968, 111, 277. (9) Meerwein, H.; Delfs, D.; Morshel, H. Angew. Chem. 1960, 72, 927. (10) Han, M. J. J. Korean Chem. SOC.1978,22, 173. (11) Saegusa, T.; Ikeda, H.; Fujii, H. Polym. J. 1973,4, 87. (12) (a) Saegusa, T.; Ikeda, H.; Fujii, H. Macromolecules 1972,5, 354. (b) Saegusa,T.; Kobayashi, S.; Kimura, Y. Zbid. 1977,10, 68,236,239. (c) Saegusa, T.; Kobayashi, S.; Furukawa, J. Zbid. 1975,8, 703. Zbid. 1976,9, 728. Zbid. 1977,10, 73. (d) Saegusa, T.; Kobayashi, T.; Chow, T. Y.; Kobayashi, S. Ibid. 1979,12, 533. (13) Weinberger, M. A.; Greenhalgh, R. Can. J . Chem. 1963, 41, 1038. (14) Morris, C. J. 0. R., et al. “Separation Methods in Biochemistry”; Wiley: New York, 1976.

Molecular Weight Dependence of Lamellar Structure in Styrene/Isoprene Two-and Three-Block Copolymers George Hadziioannout and Antoine Skoulios* C N R S , Centre d e Recherches sur les Macromolecules, 67083 Strasbourg-Cedex, France. Received December 19, 1980

ABSTRACT: The lamellar structure of three homologous series of SI, SIS, and IS1 block copolymers, 50% by weight polystyrene,was studied as a function of molecular weight with low-angle X-ray diffraction. From the structural standpoint, the three-block copolymers behave identically with the corresponding two-block copolymers of half the molecular weight. Helfand’sthermodynamic predictions for the value of the lamellar thickness are in fair numerical agreement with the experimental facts; however, the observed lamellar spacing increases with molecular weight a little more steeply than predicted. Finally, the molecular area, calculated from Helfand’s theory as a function of chemical composition at constant lamellar thickness, is independent of the exact position of the interface within the lamella. The general impression is that for tweblock copolymers the arrangement of the sequences in the lamellae would be close to a double-layer arrangement. This would imply little interpenetration of the molecules in a direction normal to the interfaces.

Introduction Since their discovery, the mesomorphic phases of block copolymers have been the subject of many studies. Some of these aimed to establish the generality of the spontaneous liquid-crystalline organization observed and to describe the main structures involved. Others analyzed the dependence of the structure on various factors such as temperature, solvent concentration, composition of the copolymer, and molecular Most of the studies that considered the role of molecular weight were carried out with binary mixtures of copolymer and solvent. They showed that the dimensions of the segregated microdomains were comparable to those of the macromolecules themselves and that they increased monotonically with the length of the blocks.’B2 Their quantitative analysis is, however, complicated because it is necessary to take into account the particular role of each solvent used. Whether good or bad, the solvents can in fact be preferential to some extent for each type of block. Several studies have nevertheless been performed in the absence of solvent. Based on scattered results obtained from an incomplete and truncated series of styrene/ isoprene or styrene/ butadiene block copolymers, the conclusions are still neither very coherent nor convincing. Present address: Polymer Research Institute, University of Massachusetts, Amherst, Mass. 01003.

For instance, Vanzo6 and Kawai7 pointed out for the case of the lamellar structure that the thickness of the layers is proportional to the square root of the molecular weight. Later, Kromer8 stated that the dimensions of the segregated microdomains are proportional to the molecular weight raised to the 0.55-0.60 power. Later still, Gallotg simply announced that the structural parameters increase with the length of the blocks in a monotonic fashion. Concerning the role of the molecular structure, he added that on going from a given two-block to the related three-block copolymer (obtained from the former by the addition of a new block), the thickness of the lamellae decreases markedly. Regarding the role of the molecular weight on the occurrence and the stability of the mesomorphic phases, it has not been possible to tackle the problem experimentally. As far as we are aware, only one study has touched on this question. This studylo concerns the change in the mechanical behavior of styrene/ butadiene block copolymers which was observed above a threshold of molecular weight of about 10OOO and which was interpreted as being related to microphase separation. All previous studies, the earliest of which were made 15 years ago, have been accompanied by theoretical explanations based on therrn~dynamics.~~~J~-~~ We will return to this later. In order to try to clarify the situation, we decided to repeat the earlier work and study further styrene/isoprene

0024-9297/82/2215-0258$01.25/0 0 1982 American Chemical Society

M , Dependence of Lamellar Structure in SI Copolymers 259

Vol. 15, No. 2, March-April 1982 Table I Molecular Characteristics of Polymers sample SI-1 SI-2" SI-3" Si-4" SI-5" SI-6 SI-7b SI-8b SI-Sb SI-10 SI-11 SI-12b SIS-1" SIS-2" SIS-3" ISI-1" ISI-2" ISI-3" ISI-4"

M,+ 8% 8500 17000 27000 51000 72000 98000 114800 130300 155000 181000 185000 205000 32500 63000 96000 48500 59900 77000 134000

M,

+

5% M,/M,

10200 18800 31000 56300 81000 118300 130000 172000 190000 210000 240000 245000 41200 81000 113500 57000 73800 97000 175000

1.27 1.15 1.25 1.20 1.15 1.30 1.20 1.32 1.30 1.16 1.35 1.30 1.27 1.30 1.20 1.20 1.25 1.30 1.35

Table I1 Structural Characteristics of PoIymers

UVXPS ?

0.005

0.550 0.495 0.500 0.505 0.490 0.525 0.597 0.525 0.495 0.490 0.472 0.535 0.494 0.504 0.542 0.476 0.494 0.502 0.545

" Obtained from Ecole d'A plication des Hauts Polym>res, Strasbourg, France. gPerdeuterated styrene blocks. block copolymers in the dry state. We chose to consider only polymers of the highest possible homogeneity with regard to molecular weight distribution and chemical composition. In addition, we were careful to select samples that consisted of a homologous series of two- and threeblock copolymers. However, we limited our investigation to copolymers containing equivalent volumes of styrene and isoprene and showing therefore only lamellar structure, which is easier to analyze.

Experimental Section The polymers used were synthesized anionically in benzene at 25 "C with sec-butyllithium as the initiator.z3 Their molecular characteristics (Table I) were carefully determined following the usual physicochemical methods. Number-average molecular weights, M,, were measured with a 502 Mechrolab osmometer at 30 "C, using toluene as solvent. Weight-average molecular weights, M,, were determined a t 25 "C with a Sofica light scattering apparatus provided with a He-Ne laser source (632 nm), using tetrahydrofuran as solvent. Due to the homogeneity of the samples with regard to molecular weight distribution and chemical composition, the apparent molecular weights measuredz4 could be safely assumed to be very close to the real weight-average molecular weights that were sought. Polydispersities were calculated from the experimental M, and M, values and corroborated by gel permeation chromatography. The GPC measurements were performed a t 25 "C with a Waters instrument with Styragel columns and tetrahydrofuran as the solvent. The weight fraction of styrene in the copolymers, w x p s , was determined from the UV absorption a t 262 nm in tetrahydrofuran solutions. The X-ray diffraction experiments were carried out with a Baro-Luzzati small-angle camera.25 This was equipped with a bent quartz monochromator and was operated under vacuum using Cu Kal radiation. The diffraction patterns were recorded at room temperature on a photographic film placed 400 mm from the specimen. Samples were molded under vacuum and oriented into large single-crystalline domains using a shearing technique described elsewhere.= They were finally annealed for a long time above the glass transition temperature of polystyrene. T o be precise, it should be noted that a t the ambient temperature at which they were examined, the copolymers were, of course, not in equilibrium since the polystmne blocks were in the glassy state. However, the measured speeings remained valid since they do not differ significantly from those measured at 130 "C. From the relative intensity of the successive diffraction lines taken two by two, it was possible to measure the relative thicknesses of the polystyrene and polyisoprene sublayers and to determine

SI-1 SI-2 SI-3 SI-4 SI-5 SI-6 SI-7 SI-8 SI-9 SI-10 SI-11 SI-12 SIS-1 SIS-2 SIS-3 ISI-1 ISI-2 ISI-3 ISI-4

(111) (138) 203 321 414 595 650 715 760 860 943 988 (147) 244 320 278 268 340 439

*

0.005

xpSa

0.505 0.502 0.497 0.520 0.614 0.529 0.511

0.503 0.504 0.494 0.522 0.606 0.527 0.503 (0.490) 0.485 0.539

0.497 0.542 0.505 0.522 0.472 0.497 0.507

Si-

dPS i-

XRXPS

disample 1%,A

0.504 0.532 0.474 0.496 0.505 0.545

2%,

d',b

lo%,

A

A

A2

94 150 189 288 359 337 340 390 422 496

223 347 439 540 592 653 737 819 832 884 ~~~

114 251 158 335 111 280 1 2 3 248 159 292 239 403

454 ~. 543 595 561 570 599 673 720 672 705 441 510 300 382 387 518

" xps = ( u v x p ~+ xRxPs)/2;see uvxps in Table I. d ' = lamellar thickness calculated with Helfand's theory and numerical values of thermodynamic parameters. 1 9 - 2 a therefrom the volume fraction and also the weight fraction, xRzps (Table 11), of polystyrene in each c o p ~ l y m e r . ~ ~ - ~ ~

Determination of the Structural Parameters The X-ray patterns that we registered generally showed many, sometimes up to 15, sharp and equidistant diffraction lines indicative of a well-developed lamellar structure. The thickness, d, of the lamellae, that is, the thickness of one polystyrene and one polyisoprene sublayer superposed, was measured from the weighted average of the reciprocal spacings (d-l = Cid;l/Ci; i is the order of diffraction) of all observed reflections (Table 11). However, the patterns of copolymers SI-1, SI-2, and SIS-1 contained only one diffraction band. This will be discussed later in the section devoted to the melting of the mesophases. Let us say at the moment that we also characterized these bands (probably in an improper sense) by their "Bragg spacing" calculated from the position of the maximum in the intensity distribution. From these data and all the molecular characteristics (Table I) that we have been able to gather for the copolymers considered in this work, we have derived several extra structural parameters that are of interest in the discussion of the conformational behavior of the blocks. In particular, we have calculatedl the thickness of the polystyrene sublayers dps = d / [ l + ~ p I ( 1- xps)/~psxpsl and the molecular area which is the average surface occupied at the interfaces by each covalent junction point between styrene and isoprene blocks (Table 11). In these formulas, ijpI and ups a r e the specific volumes of polyisoprene (1.106 cm3 g-l) and polystyrene blocks (0.953 or 0.885 cm3 g-l, depending upon whether polystyrene is hydrogenated or deuterated). Mps is the number-average molecular weight of polystyrene blocks, N is Avogadro's number (6.02 X and v is the number of polyisoprene blocks linked to each polystyrene block. Finally, on close inspection of the intensity of the diffraction lines and more specifically from its overall decay

260 Hadziioannou and Skoulios

Macromolecules

I 1000 -

3.51

d = KW*

-

*U 0

500 -

o,n.$

0

100

o :

&.lo-’

M , . ~ o - ~ 200

-

Figure 1. Variation of lamellar spacing d as a function of molecular weight (M,) for the tweblock copolymers and m a function of half of it ( M , / 2 ) for the three-block copolymers.

I -

500

Figure 3. Double-logarithmic plot of the variation of the spacing d as a function of M,. ration of polystyrene in the It is worth noting that dps definitely increases with molecular weight much more quickly than predicted by Gaussian statistics of ideal coils. This is clear in Figure 3, where log d is plotted as a function of log M. Practically, d seems proportional to Ma, and the power calculated from the slope of the straight line, a = 0.79 f 0.02, is significantly larger than that (a = 1/2) predicted for ideal coils. It also is more than the values predicted by all the phenomenological thermodynamic t h e ~ r i e s ? - ~ which J ~ J ~ consider explicitly the entropy of the polymer chains. Correspondingly, the molecular area S (Table 11) increases weakly with molecular weight ( S Mo.21),particularly when the latter is important. The nonideal behavior of the blocks is, of course, not surprising since the blocks are incompatible with one another (the interfaces are sharp) and are therefore subjected to interactions that are rather strong. Curve 2 of Figure 2 corresponds to the predictions of Leibler’s t . h e ~ r y This . ~ ~ is a microscopic statistical theory of phase equilibria in noncrystalline block copolymers. We will return to this later when we discuss the melting of the mesophases. Without going into detail regarding the method used for the calculations, let us say simply that for a lamellar two-block copolymer containing equal amounts of polystyrene and polyisoprene, the theory predicts that near the phase transition leading from the disordered to the lamellar state, the thickness of the lamellae will be d = TR,where R is the radius of gyration of the copolymer chains, assumed to be ideal. As is proper, theory is found to agree with experiment (Figure 2) only in the region of low molecular weights, where, as we will see later, the ordering transition of the system occurs. Curve 1 of Figure 2 represents the thickness dpS as calculated with Helfand’s theory.1*22 Assuming that the interface between segregated microdomains is thin compared to the total thickness of the layers, in accordance with the experimental facts, this statistical mechanical theory solves the problem of the uniform density distribution of the segments by applying a self-consistent-field method. The only parameters needed are taken from experimental data in the literature such as the interaction parameter x of the styrene and isoprene units, the length of Kuhn’s statistical elements, etc. As has already been observed by Helfand on grounds of scattered experimental data, the agreement of theory with experiment is good. The discrepancy between calculated and measured thicknesses hardly exceeds 10%. This is all the more significant because the theoretical calculations do not contain any adjustable parameters, as is generally the case for all the other phenomenological thermodynamic theories presented

-

Figure 2. Variation of the thickness of the polystyrene sublayers as a function of the molecular weight of the corresponding blocks: curve 1 is calculated from the theor of Helfand;1+22 curve 2 is calculated from the theory of Leibler; curve 3 shows the variation of the root-mean-squareend-to-enddistance of ideal polystyrene

x

homopolymers in the bulk.

in reciprocal space, it was possible to appraise the width of the interfacial region.27 We noted that this remained well below 50 A, whatever the molecular weight of the sample. This was in good agreement with the value (-30 A) observed experimentally by Kawai30 or calculated theoretically by Helfand.19v20

Variation of the Structural Parameters Figure 1 represents the lamellar spacing d plotted as a function of the total molecular weight (M,) for the twoblock copolymers or as a function of half of it ( M , / 2 ) for the three-block copolymers. Quite remarkably, all the data fall, within the accuracy of the measurements, on a single curve passing through the origin and increasing monotonically with molecular weight. Clearly, the joining of two two-block copolymer molecules by the ends of their polystyrene or polyisoprene blocks, transforming them into three-block copolymer molecules with the same chemical composition but with double the molecular weight, involved no detectable change in the thickness of the layers.33 Since the chain conformation seems to be unaffected, the corresponding entropic contribution of the coils presumably does not play a specific role in the overall stability of the system and in the fixing the structural parameters. Figure 2 shows how the thickness dPS (Table 11) depends upon the molecular weight of the polystyrene blocks. By way of comparison, we plotted in the same figure the variation of the root-mean-square end-to-end distance (h2)1/2 of ideal polystyrene chains as a function of molecular weight (curve 3). The data used were obtained from a low-angle neutron scattering study of the radius of gy-

Vol. 15, No.2, March-April 1982

Figure

4.

Electron micrograph of the disordered phase of co-

M. Dependence of Lamellar Structure in SI Copolymers 261

Figure 5. Electron micrograph of the lamellar phase of copolymer

polymer SIS-1.

SIS-2.

in the literature. It should be noted, however, that, expressed as a power law, Helfand's numerical results (see Table II) lead to an exponent (a= 0.64) that is much lower than that obtained experimentally by us (a N 0.79).

theories have tried t o treat the observed mesomorphic phases by considering the presence of the covalent bonds linking the blocks at the interface^'"'^ hut they have led to similar conclusions. In a recent microscopic statistical theory of phase equilibria in amorphous two-block copolymers?*the onset of ordering was interpreted in terms of only two relevant parameten: fmt, the product X Nof the Flory x parameter characterizing the repulsive interaction of the two types of monomer and the degree of polymerization N and second, the chemical composition of the system. It was shown that ordering occurs when x N exceeds a threshold value that dependsonly on the chemical composition of the comlvmer. Thus. for a cooolvmer . _ with a eiven chemical composition a t a fixed temperature (x is constant), organization should occur for N lying above a well-defined value.

Melting of the Mesophase All of the X-ray diffraction patterns that we recorded showed from 3 to 15 sharp Bragg reflections, indicating a well-developed lamellar organization. As we have said in the section devoted to the determination of the structural parameters, the patterns of copolymers SI-1, SI-2, and SIS-1 contain only one diffraction band which, in addition, is relatively large. The difference between the two tvoes of Dattem is strikine. T a k i i into consideration the Gct thai this abrupt change occ& as a function of molecular weight at a b u t 20000 for two-block and 40000 for three-block copolymers (see Figure l), one is tempted to speak of a phase transition between the well-organized lamellar phase and a disordered phase. Electron microscopy of copolymers SIS-1 and SIS-2, whose molecular weights are similar (Table I) but lie on both sides of the molecular weight threshold mentioned above, fully corroborates this interpretation (Figures 4 and 5). It clearly shows that SIS-2, whose X-ray diffraction pattern is formed of three very sharp Bragg reflections, is lamellar in structure. Furthermore, SIS-1, whose X-ray pattern contains only one broad diffraction band, is actually in the disordered state. Its structure corresoonds to a simple interdispersion of nodular polystyrene-and polyisoprene microdomains. Based on this observation, the change in mechanical behavior reported by Bishoplo for styrene/ butadiene block copolymenr is no doubt due to the ordering and to the establishment of a true oeriodic arrangement and not only (as claimed) to the segregation of theblocks into distinct microphases. As far as the microphase separation is concerned, it is easy to understand the role of molecular weight. Segregation in a binary polymer mixture is merely due to the decrease of entropy of mixing with increasing degree of polymerization. Assuming that the segregation of the blocks in a copolymer is not very different from the segregation of the corresponding homopolymers in a binary mixture, Fedorsl6 has shown by simple thermodynamic calculations based on the Flory-Huggins theory that the segregation in stytenelisoprene block copolymers, and the ordering itself, should occur for molecular weights lying in the range from 10000 to 20000. Quite surprisingly, the result is nicely in agreement with the facts, in spite of the unrealistic assumptions made. Other thermodynamic

. -

-

Discussion It is interesting to note that, within their respective ranges of applicability, the two theories mentioned above account quantitatively for the structural parameters observed. Helfands theory applies over the whole range of molecular weights considered because the interfaces have indeed negligible thickness almost all the way down to the melting threshold and Leibler's theory is valid in the low molecular weight range, where, as we have seen, the ordering Drocess b e ~ n to s occur. Furthermore, it should be state2 ihat both?heories admit implicitly that the conformation of the chains differs from the ideal. However. they do not, at least for the time being, expand on this deviation, which nevertheless seems very large. In order to obtain information on this point. we have proceeded in the following indirect way. Fiom the theory of Helfandl*cn we have calculated for a styrene/isoprene copolymer having an overall molecular weight of about 1OOooO the molecular area S as a function of the weight fraction xps of polystyrene. For each composition we have simply corrected the molecular weight in such a way that the overall thickness of the lamellae remains equal to 600 A. The result suggests (Table 111) that the molecular area S remains independent of the composition. It seems that the lateral space occupied by the junction points between blocks does not depend on the ratio d,/d, which is the exact position of the interface within the lamellae. The polymer chains seem to keep the same lateral bulkiness a t any distance from the interface. This bulkiness is presumably determined almost entirely by small-scale interactions located in the interfacial region.

262 Hadziioannou and Skoulios

Macromolecules

Table I11 Variation of the Molecular Area as a Function of the Composition for a Lamellar System of Thickness 600 A Calculated b y Helfand’s XPS

0.2 0.3 0.4 0.5 0.6 0.7 0.8

Mn 96 000 97 000 98000 100@00

103 000 lOGOOO 110000

dPdd 0.177 0.269 0.364 0.462 0.563 0.667 0.775

s, A’ 569 567 566 568 576 585 598

In summary, we have first shown that from a structural standpoint three-block copolymers behave identically with the corresponding two-block copolymers of the same chemical composition but of half the molecular weight. It seems therefore that the detailed conformation of the chains on a macromolecular scale and as a consequence the entropy of the blocks does not play a decisive role in the stability of the system and in the determination of the values of the geometrical parameters characterizing the mesophase. It was then shown that as a first approximation the lamellar spacing is proportional to the molecular weight raised to the 0.79 power. This exponent is much higher than that (1/2) predicted for ideal random coils and still higher than the exponent (0.64) that would numerically fit Helfand’s theory. Its value is not excessively far from unity, which means that the structural parameters are not far from being fully determined, on the one hand, by the molecular area of the molecules at the interfaces, and on the other (in a trivial way), by their molar volume. It was finally shown that, for a given total lamellar thickness, the molecular area is consistently independent of the exact position of the interface in the layers. As a whole, this information suggests that everything behaves like sequences in two-block copolymers, hardly interpenetrating one another in a direction normal to the interfaces and adopting within each sublamella a type of double-layer configuration.

Acknowledgment. We thank Dr. Y. Gallot for help with the synthesis of the deuterated polymers, Dr. B. Lotz for help with the electron microscopy observations, Ecole d’Application des Hauts Polym6res for supplying many polymers, Dr. L. Leibler for giving us a preprint of his article,32and Dr. A. Hodge for help with the translation to English. G.H. is grateful to the French Government for financial support with a foreign postgraduate student fellowship. References and Notes (1) Skoulios, A. Adu. Liq. Cryst. 1975,1, 169. (2) Gallot, B. Adu. Polym. Sci. 1979,29, 85. (3) Molau, G.E. “Block Copolymers”; Aggarwal, S. L., Ed.; Plenum Press: New York, 1970;p 79.

(4) Folkes, M. J.; Keller, A. ‘Physics of Glassy Polymers”; Haward, R. N., Ed.; Applied Science Publishers: London, 1973; p 548. (5) Gallot, B. “Liquid Crystalline Order in Polymers”; Blumstein, A., Ed.; Academic Press: New York, 1978; p 191. (6) Bradford, E.; Vanzo, E. J. Polym. Sci., Part A-1 1968,6,1661. (7) Inoue, T.; Soen, T.; Hashimoto, T.; Kawai, H. J. Polym. Sci., Part A-2 1969,7,1283. ‘Block Copolymers”; Aggarwal, S. L., Ed.; Plenum Press: New York, 1970;p 53. (8) Kromer, H.; Hoffman, M.; Kampf, G. Ber. Bunsenges. Phys. Chem. 1970,74,859. (9) Douy, A,; Gallot, B. Makromol. Chem. 1972,156,81. (10) Holden, G.; Bishop, E. T.; Legge, N. R. J. Polym. Sci., Part C 1969,26, 37. (11) Bianchi, U.;Pedemonte, E.; Turturro, A. Polymer 1970,11, 268. (12) Leary, D. F.; Williams, M. C. J . Polym. Sci., Part B 1970,8, 335. (13) La Flair, R.T. Pure Appl. Chem. 1971,8,195. (14) Meier, D. J. J. Polym. Sci., Part C 1969,26,81.Polym. Prepr., Am. Chem. SOC.,Diu. Polym. Chem. 1970,11,400. (15) Krause, S.J.Polym. Sci., Part A-2 1969,7, 249. Macromolecules 1970,3,84.“Block and Graft Copolymers”;Burke, J. J., Weiss, V., Eds.; Syracuse University Press: Syracuse, N.Y., 1973;p 143. (16) Fedors, R.F. J . Polym. Sci., Part C 1969,26, 189. (17) Krigbaum, W. R.;Yazgan, S.; Tolbert, W. R. J. Polym. Sci., Polym. Phys. Ed. 1973,11, 511. (18) SvmD . . Boehm. B. R.:Kriebaum. W. R. J. Polvm. Sci.. Polvm. “ 1976,No. 54,’153.(19) Helfand. E.: Taeami. Y. J. Polvm. Sci.. Polvm. Lett. Ed. 1971. 9, 741. J. C h e k Phys. 1972,-56,3592. Ibid. 1972,57,1812: (20) Helfand, E. Polym. Prepr., Am. Chem. SOC.,Diu. Polym. Chem. 1974,15,246. J. Chem. Phys. 1975,62, 999. Macromolecules 1975,8,552. (21) Helfand. E.: Wasserman. Z. R. Macromolecules 1976.9.879. ’ Polym. Eng. Sci. 1977,17,582. (22) Helfand. E.: Same. A. M. J. Chem. Phvs. 1975.62. 1327. (23) Szwarc, ’ M.’ “C!arbanions, Living PoGmers and ’ Electron Transfer Process”; Interscience: New York, 1968. (24) Leng, M.; Benoit, H. J . Chim. Phys. (Paris) 1961,58, 480. (25) Luzzati, V.; Baro, R. J.Phys. Radium 1961,22, 186A. (26) Mathis, A,; Hadziioannou, G.; Skoulios, A. Polym. Prepr., Am. Chem. SOC.,Diu. Polym. Chem. 1977,18(l),282. Polym. Eng. Sci. 1977,17,570. Colloid Polym. Sci. 1979,257, 136. (27) Skoulios, A. ‘Block and Graft Copolymers”; Burke, J. J., Weiss, V., Eds.; Syracuse University Press: Syracuse, N.Y., 1973;p 121. (28) Ptaszynski, B.; Terrisse, J.; Skoulios, A. Makromol. Chem. 1975,176,3483. (29) Ionescu, M. L.; Skoulios, A. Makromol. Chem. 1976,177,257. (30) Hashimoto, T.; Todo, A,; Itoi, H.; Kawai, H. Macromolecules 1977,10,377. (31) Cotton, J. P.; Decker, D.; Benok, H.; Farnoux, B.; Higgins, J.; Jannink, G.; Ober, R.; Picot, C.; des Cloizeaux, J. Macromolecules 1974,7, 863. (32) Leibler, L. Macromolecules 1980,13, 1602. (33) This property has already been used in the form of an assumption by Uchida and then by Helfand (see: Uchida, T.; Soen, T.; Inoue, T.; Kawai, H. J. Polym. Sci., Part A-2 1972, 10, 101; see also ref 21) for the purpose of simplifying the theoretical thermodynamic discussion of block copolymers. However, this assumption was considered (see Uchida et al. on p 109)only as “a crude but simple expedient...,so that micelles of two- and three-block copolymers could...be discussed to the same order of approximation by a relatively simple thermodynamic approach.” It is of interest to note that, in fact, this assumption turns out to be conspicuously realistic. I

I~